U.S. patent number 6,983,326 [Application Number 09/921,940] was granted by the patent office on 2006-01-03 for system and method for distributed function discovery in a peer-to-peer network environment.
This patent grant is currently assigned to Networks Associates Technology, Inc.. Invention is credited to Martin Fallenstedt, Victor Kouznetsov, Daniel Melchione, Charles L. Vigue.
United States Patent |
6,983,326 |
Vigue , et al. |
January 3, 2006 |
**Please see images for:
( Certificate of Correction ) ** |
System and method for distributed function discovery in a
peer-to-peer network environment
Abstract
A system and method for distributed function discovery with
third party responses in a peer-to-peer network to facilitate
efficient use of bandwidth and resources are disclosed. The method
for facilitating distributed function discovery in a peer-to-peer
network generally comprises receiving a broadcast request for a
service function from a peer client at a peer server, locating
information regarding a location remote to the peer server having
the requested service function using a stored list of service
functions locally stored at the peer server, and responding to the
peer client with a response containing the location remote to the
peer server if information on the requested service function is
located.
Inventors: |
Vigue; Charles L. (Lapine,
OR), Fallenstedt; Martin (Beaverton, OR), Melchione;
Daniel (Beaverton, OR), Kouznetsov; Victor (Aloha,
OR) |
Assignee: |
Networks Associates Technology,
Inc. (Santa Clara, CA)
|
Family
ID: |
35509154 |
Appl.
No.: |
09/921,940 |
Filed: |
August 2, 2001 |
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
60298681 |
Jun 15, 2001 |
|
|
|
|
60282333 |
Apr 6, 2001 |
|
|
|
|
Current U.S.
Class: |
709/229;
707/999.104; 707/999.107; 713/168; 717/103; 719/330 |
Current CPC
Class: |
H04L
63/12 (20130101); H04L 67/104 (20130101); H04L
67/1063 (20130101); Y10S 707/99945 (20130101); Y10S
707/99948 (20130101) |
Current International
Class: |
G06F
15/16 (20060101); G06F 3/00 (20060101); G06F
7/00 (20060101); G06F 9/44 (20060101); H04L
9/00 (20060101) |
Field of
Search: |
;709/219,229 ;707/104.1
;717/103 ;713/168 ;719/330 |
References Cited
[Referenced By]
U.S. Patent Documents
|
|
|
5864871 |
January 1999 |
Kitain et al. |
6212633 |
April 2001 |
Levy et al. |
6615253 |
September 2003 |
Bowman-Amuah |
6697824 |
February 2004 |
Bowman-Amuah |
6782527 |
August 2004 |
Kouznetsov et al. |
6820199 |
November 2004 |
Wheeler et al. |
6850985 |
February 2005 |
Giloi et al. |
6851053 |
February 2005 |
Liles et al. |
|
Other References
Kovalerchuck et al., Comparison of relational methods and
attribute--based methodss for data mining in intelligent systems,
proceedings of the 1999 IEEE, International Symposium on
Intelligent System and Semiotics, Cambridge, MA, pp. 162-166, Sep.
1999. cited by examiner .
Zhao et al., A Scalable Wireless Virtual LAN, Nov. 1996, Proceeding
of the 2nd annual international Conference On Mobile Computing and
Networking. cited by examiner .
Cameron R. Using Mobile Agents for Network Resource Discovery in
Peer-toPeer Networks, Jun. 2001 ACM SIGecom Exchanges, vol. 2,
issue 3. cited by examiner .
Eugster et al., Lightweight Probabilistic Broadcast, Dependable
Systems And Networks, 2001, Proceeding, The International
Conference on Jul. 1-4, 2001, pp. 443-452. cited by
examiner.
|
Primary Examiner: Harvey; Jack
Assistant Examiner: Lin; Kelvin
Attorney, Agent or Firm: Zilka-Kotab, PC Hamaty; Christopher
J.
Parent Case Text
CROSS REFERENCE TO RELATED APPLICATIONS
This application claims the priority benefit of U.S. Provisional
Patent Application No. 60/282,333, entitled "System and Method for
Efficient Use of Bandwidth and Resources in a Peer-to-Peer Network
Environment" and filed Apr. 6, 2001 and U.S. Provisional Patent
Application No. 60/298,681, entitled "System and Method for
Efficient Updating of Virus Protection Software and Other Efficient
Uses of Bandwidth and Resources in a Peer-to-Peer Network
Environment" and filed Jun. 15, 2001, both of which are
incorporated herein by reference in their entireties.
Claims
What is claimed is:
1. A method for facilitating distributed function discovery in a
peer-to-peer network, comprising: receiving a broadcast request for
a service function from a peer client at a peer server; locating
information regarding a location remote to the peer server having
the requested service function using a stored list of service
functions locally stored at the peer server; and responding to the
peer client with a response containing the location remote to the
peer server if information on the requested service function is
located; wherein said peer server listens for a broadcast response
packet from the peer client over the network for a randomly
generated delay response period prior to said responding, wherein
said responding is only performed upon non-receipt of the response
packet at expiry of the delay response period, and said responding
is cancelled upon receipt of the broadcast response packet during
the randomly generated delay response period.
2. A method for facilitating distributed function discovery of
claim 1, wherein the response is digitally signed.
3. A method for facilitating distributed function discovery of
claim 2, wherein the digitally signed response is signed by a
1024-bit VeriSign digital certificate.
4. A method for facilitating distributed function discovery of
claim 1, further comprising: receiving a packet regarding a
remotely located designated service function provider; and storing
information regarding the remotely located designated service
function provider.
5. A method for facilitating distributed function discovery of
claim 1, wherein the randomly generated delay ensures that
responses performed by a plurality of the peer servers are
distributed among the peer servers.
6. A method for facilitating distributed function discovery of
claim 1, wherein the broadcast request includes the following
format: <service type="X" version="X" ID="X" method="X"
href=http://X acceptprotoco="X".
7. A method for distributed function discovery in a peer-to-peer
network, comprising: broadcasting a packet requesting a service
function; receiving a response from a responding peer server, the
packet containing information regarding a designated provider for
the requested service function, the information including a
location of the designated provider remote to the responding peer
server; and accessing the requested service function from the
designated service provider at the location specified in the
response of the responding peer server; wherein said peer server
listens for a broadcast response packet from a peer client over the
network for a randomly generated delay response period prior to
said responding, wherein said response is only performed upon
non-receipt of the response packet at expiry of the delay response
period, and said responds is cancelled upon receipt of the
broadcast response packet during the randomly generated delay
response period.
8. A method for distributed function discovery in a peer-to-peer
network of claim 7, wherein the response is digitally signed.
9. A method for distributed function discovery in a peer-to-peer
network of claim 8, wherein the digitally signed response is signed
by a 1024-bit VeriSign digital certificate.
10. A method for distributed function discovery in a peer-to-peer
network of claim 7, wherein the randomly generated delay ensures
that responses performed by a plurality of the peer servers are
distributed among the peer servers.
11. A method for distributed function discovery in a peer-to-peer
network of claim 7, wherein the packet includes the following
format: <service type="X" version="X" ID="X" method="X"
href=http://X acceptprotoco="X".
12. A computer program product for facilitating distributed
function discovery in a peer-to-peer network, comprising: computer
code that receives a broadcast request for a service function from
a peer client at a peer server; computer code that locates
information regarding a location remote to the peer server having
the requested service Junction using a stored list of service
functions locally stored at the peer server; computer code that
responds to the peer client with a response containing the location
remote to the peer server if information on the requested service
function is located; and a computer readable medium that stores
said computer codes; wherein said peer server listens for a
broadcast response packet from the peer client over the network for
a randomly generated delay response period prior to said
responding, wherein said responding is only performed upon
non-receipt of the response packet at expiry of the delay response
period, and said responding is cancelled upon receipt of the
broadcast response Packet during the randomly generated delay
response period.
13. A computer program product for facilitating distributed
function discovery of claim 12, wherein the response is digitally
signed.
14. A computer program product for facilitating distributed
function discovery of claim 13, wherein the digitally signed
response is signed by a 1024-bit VeriSign digital certificate.
15. A computer program product for facilitating distributed
function discovery of claim 12, further comprising: computer code
that receives a packet regarding a remotely located designated
service function provider; and computer code that stores
information regarding the remotely located designated service
function provider.
16. A computer program product for facilitating distributed
function discovery of claim 12, wherein the randomly generated
delay ensures that responses performed by a plurality of the peer
servers are distributed among the peer servers.
17. A computer program product for facilitating distributed
function discovery of claim 12, wherein the broadcast request
includes the following format: <service type="X" version="X"
ID="X" method="X" href=http://X acceptprotoco="X".
18. A computer program product for distributed function discovery
in a peer-to-peer network, comprising: computer code that
broadcasts a packet requesting a service function; computer code
that receives a response from a responding peer server, the packet
containing information regarding a designated provider for the
requested service function, the information including a location of
the designated provider remote to the responding peer server;
computer code that accesses the requested service function from the
designated service provider at the location specified in the
response of the responding peer server; and a computer readable
medium that stores said computer codes; wherein said peer server
listens for a broadcast response packet from a peer client over the
network for a randomly generated delay response period prior to
said responding, wherein said response is only performed upon
non-receipt of the response packet at expiry of the delay response
period, and said response is cancelled upon receipt of the
broadcast response packet during the randomly generated delay
response period.
19. A computer program product for distributed function discovery
of claim 18, wherein the response is digitally signed.
20. A computer program product for distributed function discovery
of claim 19, wherein the digitally signed response is signed by a
1024-bit VeriSign digital certificate.
21. A computer program product for distributed function discovery
of claim 18, wherein the randomly generated delay ensures that
responses performed by a plurality of the peer servers are
distributed among the peer servers.
22. A computer program product for distributed function discovery
of claim 18, wherein the packet includes the following format:
<service type="X" version="X" ID="X" method="X" href=http://X
acceptprotoco="X".
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to a system and method for
efficient use of bandwidth and resources in a peer-to-peer network
environment. More specifically, a system and method for distributed
function discovery with third party responses in a peer-to-peer
network to facilitate efficient use of bandwidth and resources are
disclosed.
2. Description of Related Art
Conventionally, to obtain anti-virus product updates and/or
signature files, computers rely on a pull approach in which each
client or server computer must retrieve the updated anti-virus file
directly from a source via the Internet. For a computer network, a
network administrator may allow anti-virus signature files to
become out of date because there are simply too many clients on the
network for effective management. Alternatively, the network
administrator may schedule the clients to automatically pull the
updated anti-virus file from the Internet when each client logs
onto the computer. However, such an approach can result in a
bandwidth crunch such as in the early morning work hours when most
users log onto their computers.
Connections to the Internet from within an organization,
particularly from a small to medium sized organization, may be
relatively slow. For example, a small to medium sized business may
share a single cable or DSL modem, a 56K modem, or an ISDN line. In
contrast, in a typical work group interconnected via a LAN,
connections on the LAN are generally much faster, the typical LAN
being 100/TX (100 Mbps). Peer-to-peer networks thus partially
address the need for efficient use of bandwidth and resources in a
computer network.
However, an actual service provider in a peer-to-peer network may
not be easily locatable by other peers in the network. In
particular, because there is no single dedicated server in a
peer-to-peer network, a node attempting to locate a service
provider may not be able to locate the desired service provider or
may unnecessarily expend efforts and utilize resources to locate
the desired service provider.
Thus, it is desirable to provide a system and method for an
efficient and effective way to discover service providers in a
peer-to-peer network environment, particularly those service
providers that are not easily locatable.
SUMMARY OF THE INVENTION
A system and method for distributed function discovery with third
party responses in a peer-to-peer network to facilitate efficient
use of bandwidth and resources are disclosed. The peering service
system and method facilitate in spreading load amongst peers in a
distributed network interconnected via a LAN in a smooth, secure
and scalable way. The service provider selection from among the
peers is preferably achieved through an automatic selection process
using signed certificates to authenticate the legitimacy of each
potential service provider. Upon completion of the selection
process, the selected service provider optionally transmits a
broadcast message over the network to notify all other peers of the
outcome of the selection process.
It should be appreciated that the present invention can be
implemented in numerous ways, including as a process, an apparatus,
a system, a device, a method, or a computer readable medium such as
a computer readable storage medium or a computer network wherein
program instructions are sent over optical or electronic
communication lines. Several inventive embodiments of the present
invention are described below.
According to a preferred embodiment, a method for facilitating
distributed function discovery in a peer-to-peer network generally
comprises receiving a broadcast request for a service function from
a peer client at a peer server, locating information regarding a
location remote to the peer server having the requested service
function using a stored list of service functions locally stored at
the peer server, and responding to the peer client with a response
containing the location remote to the peer server if information on
the requested service function is located.
The method may further comprise listening for a broadcast response
packet over the network for a randomly generated delay response
period prior to the responding. In addition, the responding is
performed only upon non-receipt of the response packet at expiry of
the delay response period or the responding is canceled upon
receipt of the broadcast response packet during the randomly
generated delay response period. Preferably, the response is,
digitally signed, such as by a 1024-bit VeriSign digital
certificate.
The method may also include receiving a packet regarding a remotely
located designated service function provider and storing
information regarding the remotely located designated service
function provider.
According to another preferred embodiment, a method for distributed
function discovery in a peer-to-peer network generally comprises
broadcasting a packet requesting a service function, receiving a
response from a responding peer server, the packet containing
information regarding a designated provider for the requested
service function, the information including location of the
designated provider remote to the responding peer server, and
accessing the requested service function from the designated
service provider at the location specified in the response of the
responding peer server.
The method may include verifying that the response election packet
wins over the initiating election packet based on the value for the
criteria. In addition, the election result broadcast may also
contain a value for the criteria such that verifying the election
result includes verifying that the value for the criteria in the
response election packet wins over the value for the criteria in
the initiating election packet. The response time-out period is
optionally at least a sum of maximum delay election response period
and round trip transmission time.
Preferably, the method includes verifying a digital signature of
the response election packet and the election result broadcast.
Each of the digitally signed election initiating packet and
digitally signed election result packet may be signed by a 1024-bit
VeriSign digital certificate.
According to another preferred embodiment, a method for secure
automatic selection of a designated service provider in a
peer-to-peer network generally comprising receiving a digitally
signed election initiating packet containing a value for a criteria
by a receiving node from a sending node, determining one of the
nodes as current winner by comparing the values of the criteria in
the election initiating packet and at the receiving node, awaiting
for, verifying, and storing election result in an election result
broadcast if the sending node is the current winner, or awaiting
expiry of response delay period or receipt of an additional
election packet if the receiving node is the current winner, and
broadcasting a digitally signed election result packet indicating
the receiving node is the designated service provider if expiry of
response delay period occurs prior to receipt of any additional
election packet. Preferably, the response delay period is randomly
generated within a predetermined range.
These and other features and advantages of the present invention
will be presented in more detail in the following detailed
description and the accompanying figures which illustrate by way of
example the principles of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention will be readily understood by the following
detailed description in conjunction with the accompanying drawings,
wherein like reference numerals designate like structural elements,
and in which:
FIG. 1 is a block diagram of an exemplary computer network suitable
for implementing a peering service in a peer-to-peer network to
facilitate efficient use of bandwidth and resources;
FIG. 2 is a block diagram illustrating an exemplary peering service
system and method implemented at a node of the computer network of
FIG. 1;
FIG. 3 is a state diagram illustrating states of a typical peering
service server in processing a request from a peering client in the
peer-to-peer network;
FIGS. 4A and 4B are alternative state diagrams illustrating states
of a typical peering service client in requesting a resource over
the peer-to-peer network;
FIG. 5 is a flowchart illustrating a typical process of a peering
service server in processing a request from a peering client in the
peer-to-peer network;
FIG. 6 is a flowchart illustrating a typical process of a peering
service client in requesting a resource over the peer-to-peer
network;
FIG. 7 is a flowchart illustrating a preferred embodiment of the
retrieve step of FIG. 6 in more detail;
FIG. 8 is a flowchart illustrating an exemplary process implemented
by a node for storing or caching designated service provider
information as part of a distributed function discovery process in
a peer-to-peer network;
FIG. 9 is a flowchart illustrating an exemplary process implemented
by a node for serving stored or cached designated service provider
information as part of a distributed function discovery process in
a peer-to-peer network;
FIG. 10 illustrates an example of a computer system that can be
utilized with the various embodiments of method and processing
described herein; and
FIG. 11 illustrates a system block diagram of the computer system
of FIG. 10.
DESCRIPTION OF SPECIFIC EMBODIMENTS
A system and method for distributed function discovery with third
party responses in a peer-to-peer network to facilitate efficient
use of bandwidth and resources are disclosed. The peering service
facilitates in spreading load amongst peers in a distributed
network interconnected via a LAN in a smooth, secure and scalable
way. A service or an application that is service-enabled may
minimize or reduce the usage of, for example, Internet bandwidth by
attempting to locate a local aliased copy of a requested resource
residing within the peer-to-peer network. If a local aliased copy
of the requested resource is located, the requesting computer may
obtain the requested resources locally. Once the requesting
computer obtains a copy of the requested resource, whether locally
or remotely, the requesting computer may itself become a server for
the aliased copy for subsequent requests for the resource.
The following description is presented to enable any person skilled
in the art to make and use the invention. Descriptions of specific
embodiments and applications are provided only as examples and
various modifications will be readily apparent to those skilled in
the art. The general principles defined herein may be applied to
other embodiments and applications without departing from the
spirit and scope of the invention. Thus, the present invention is
to be accorded the widest scope encompassing numerous alternatives,
modifications and equivalents consistent with the principles and
features disclosed herein. For purpose of clarity, details relating
to technical material that is known in the technical fields related
to the invention have not been described in detail so as not to
unnecessarily obscure the present invention.
FIG. 1 is a block diagram of an exemplary computer network 100
suitable for implementing the peering service in a peer-to-peer
network to facilitate efficient use of bandwidth and resources as
described herein. In particular, the computer network 100 comprises
nodes, computers, or workstations 104A E interconnected via a LAN
102. It is to be understood that the LAN 102 may be implemented
using any suitable network mechanism including wire and wireless.
In the exemplary computer network 100, only two of the nodes 104D
and 104E have access to the Internet.
FIG. 2 is a block diagram illustrating an exemplary peering service
system and method implemented at a node of the computer network of
FIG. 1. As shown, each node 104 provides the functionality of both
a server 106 and a client 110. The peering service system utilizes
a port, such as port 1967, for transmitting directed or broadcast
messages to peers on the network. The server preferably includes an
embedded HTTP server 108, typically a micro HTTP server. The HTTP
server 108 allows aliased URLs to be accessed by other peers on the
network. The HTTP server 108 preferably uses an obscure port such
as port 6515 and is preferably restricted to operations required to
facilitate distribution of, for example, cached files and uploading
of data or requests.
Typically, each node runs both the server and the client. However,
each node may run only the client or the server. The peering system
and method are preferably implemented as a peering service
application ("service" or "service-enabled application") or daemon
process. It is noted that a service-enabled application need not be
a service application. For example, a service-enabled application
may also refer to a service-aware application that fires up,
communicates with the service and then shuts down
interactively.
The peering system preferably provides a linkable client API
library 112 to facilitate communication between the peering service
and any service-enabled applications. In one preferred embodiment,
the peering system may export the client API library 112 to any
service-enabled application such that the service-enabled
application may utilize the peering service to discover any type of
resource that can be identified with a URL or URI, for example.
Alternatively, a given application and the peering service may be
tightly coupled so as to eliminate the need for the linkable client
API library.
FIG. 3 is a state diagram illustrating states 120 of a typical
peering service server in processing a given request from a peering
client in the peer-to-peer network. Initially, the service server
is in an idle state 122 while listening on a designated port for a
broadcast request message from a peer client on the network. When
the service server receives a broadcast request message such as an
"I need" packet from a peering client on the network, the service
server transitions to a locating local aliased copy state 124. In
particular, the service server refers to its list of local aliased
copies to determine if the service server has a local copy of the
requested resource or item identified by, for example, an URL/URI.
If the service server determines that it does not have a local copy
of the requested resource, then the service server returns to the
server idle state 122.
Alternatively, if the service server determines that it has a local
copy, the service server preferably waits a randomly generated
delay response time period at stage 126. The service server may
generate a random number, such as between 0 and 2000, which the
service server utilizes as the length of time it waits before
responding. In one preferred embodiment, the random number is the
number of milliseconds the service server waits before replying to
the request. While the service server awaits expiry of the randomly
generated delay response time period, the service server listens
for a broadcast "I found" packet from the requesting client
corresponding to the received request packet. It is noted that
regardless of the state of the service server for a given peer
request, the service server listens for new requests such as "I
need" packets. The broadcast "I found" packet from the requesting
client indicates that the requesting client has found the requested
resource. If the service server receives an "I found" packet from
the requesting client before expiry of the delay response time
period, the service server transitions to state 128 to cancel the
response to the request and returns to server idle state 122.
Alternatively, if no "I found" packet is received prior to the
expiration of the delay response time period, the service server
transitions to state 130 to transmit an "I have" packet directly to
the requesting peer client. The "I have" packet preferably contains
a local alias for the requested object on the service server which
the requesting peer can access via the HTTP server of the of the
service server. Although not preferred, the service server may
alternatively broadcast the "I have" packet over the network rather
than transmitting it directly to the requesting client. The service
server then returns to the server idle state 122.
As is evident, the randomly generated delay response time period
allows multiple peer servers to share loads in an orderly fashion.
In particular, randomizing the delay response time period ensures
that any given node would not automatically become the default
server to a large portion of the peers and eliminates any need for
the service server to exercise preferences as to which service
clients the service server will supply the requested item. In other
words, the random wait time before responding to a request ensures
that any one machine does not become an overloaded server of the
item or update to the rest of the network. In addition, as a given
item is propagated through the network to peers on the network, the
load on any one node is likely further reduced. Thus, the system
impact on a given service server as it supplies the requested item
to service clients can be relatively minimal.
However, it is to be understood that a situation in which multiple
service servers each transmitting an "I have" packet in response to
a given request packet may occur. For example, a first service
server may transmit an "I have" packet upon expiry of its delay
response time period. The "I found" packet transmitted or to be
transmitted by the requesting peer corresponding to the first "I
have" packet may not arrive at the second service server prior to
the expiry of its delay response time period, causing the second
service server to transmit an "I have" upon expiry of its delay
response time period. In such a situation where the requesting
client receives multiple "I have" packets from multiple service
servers, the requesting client may simply process the first "I
have" response and ignore any subsequent "I have" packets it may
receive.
FIGS. 4A and 4B are alternative state diagrams illustrating states
140, 140A of a typical peering service client in making a given
request for a resource over the peer-to-peer network. Referring to
FIG. 4A, initially, the service client is in an idle state 142.
When a service client needs a desired resource, such as an Internet
resource, the service client generates and broadcasts an "I need"
packet over the peer-to-peer network. For example, the "I need"
request may specify an URL (e.g.,
http://something.tld/someother/thing/here), a protocol (e.g., HTTP
protocol), a desired operation (e.g., get operation), and that the
requesting peer only wants cached objects.
After broadcasting the "I need" request, the service client
transitions to a waiting for response state 144 in which the
service client awaits a maximum delay response time period plus a
transmission time period for a response from any of the service
servers. In the example above where the randomly generated delay
response time period ranges between 0 and 2000 milliseconds, the
client response waiting time period would be, for example, 2200
milliseconds to allow for a 200 millisecond transmission time
period.
If an "I have" response is received from a service server during
the client response waiting time, then the service client
transitions to state 146 and generates and broadcasts an "I found"
message over the network to inform all other peers that the desired
resource or item has been found. The service client then
transitions to the requested item found state 158. The
service-enabled application requesting the item then retrieves the
requested item from the responding service server at the location
within the network as specified in the received "I have" packet.
Generally, the service-enabled application retrieves the requested
item through the local HTTP server using, for example, port 6515.
Once the service-enabled application successfully retrieves the
requested item, the service client informs the service server
running on the same machine that a local copy of the resource now
exists. The service client then returns to the idle state 142.
Alternatively, if no response is received during the client
response waiting time, the service client times out and transitions
to state 150 to retrieve the requested item itself such as via the
Internet. Once the retrieval is complete, the service client
transitions to found item state 158 in which the service client
informs the service server running on the same computer or at the
same node that a local copy of the resource now exists. The service
client then returns to client idle state 142. As is evident,
regardless of whether the service client received an "I have"
packet from a service server on the network, the client machine can
itself become a service server for the requested resource after
successful completion of its request.
FIG. 4B illustrates the 140A states of a typical service in a
preferred alternative embodiment particularly suitable for
applications that include downloading of files. States 140A
includes the states as shown and described with reference to FIG.
4A plus additional states for dealing with currently in-progress
downloads of the requested item by other peers. These additional
states allow a peer node to complete downloading the requested
resource and then distribute it immediately and automatically upon
download completion to the requesting service client.
In particular, instead of directly transitioning to state 150 to
retrieve the requested item itself after the service client times
out the request, the service client transitions to "wait for
download?" state 144 in which the service client determines whether
it can or will wait for completion of any in-progress download of
the requested item by another peer. If not, then the service client
transitions to state 150 to retrieve the requested item itself and
continues with state transitions similar to that described above
with reference to FIG. 4A.
If the service client determines that it can or will wait for the
completion of any in-progress download, the service client
transitions to "any in-progress downloads?" state 152. If there are
no such in-progress downloads of the requested item, then the
service client transitions to state 150 to retrieve the requested
item itself and continues with state transitions similar to that
described above with reference to FIG. 4A.
Alternatively, if there is at least one in-progress download of the
requested item, then the service client transitions to state 154 in
which it generates and broadcasts an "I found" message. The service
client then transitions to state 156 to await completion of the
in-progress download of the requested item. Upon completion of the
in-progress download of the requested item, the service client
transitions to the requested item found state 158. The service
client retrieves the requested item from the local location within
the network. After successful completion of its request, the
service client will inform the service server running on the same
machine that a local copy of the resource now exists. The service
client then returns to the idle state 142.
As is evident, in order for the service client to determine if
there are any in-progress downloads in state 152, a service client
that is downloading a file for a service-enabled application
preferably broadcasts a "downloading" message and/or directly
responds to the client server of a broadcast "I need" request with
a "I am downloading" rather than an "I have" response message. In
one preferred embodiment, the service client may set a downloading
flag for the corresponding file to true.
In addition, the service-enabled application preferably transmits
periodic progress packets to any node that is waiting for the
resource being downloaded such that those nodes may interactively
display download progress information to end users at state 156.
Alternatively, the service-enabled application may broadcast such
periodic download progress packets over the network. Thus, a node
in the retrieve item state 150 preferably periodically transmits a
"downloading" message that includes progress information.
Service Functionality and Service Packet Format
One functionality provided by the peering service is that of a
central clearing house for formatting, sending, receiving and
decoding service packets, such as for "I need," "I found," and "I
have" packets. In other words, the peering service manages the
peer-to-peer communication process for obtaining requested items.
The specific functionality invoked by a given service packet itself
is generally dependent on the specific service-enabled
application.
The communication protocol used in broadcasts (e.g., "I need" and
"I found" packets) and responses (e.g., "I have" packets) is
typically TCP/IP. Each packet is typically approximately 200 bytes
in size and contains the node ID of the sender as well as any other
suitable information. Transfer of the requested item from the
service server to the service client is typically via HTTP.
The service packet format is preferably based upon the
well-accepted and widely utilized XML format. For example, an XML
service packet format may include a service identification and
various key-value pairs, including those inserted by the service as
well as those defined by the corresponding service-enabled
application.
Various key-value pairs may be inserted by the peering service into
each service packet. Examples of suitable key-value pairs include
identification, type, and version key-value pairs. Specifically, an
identification key-value pair identifies each request and responses
corresponding to the request. In general, the identification value
is unique on the originating node but need not be unique on the
network as a whole. The range of values for the identification
value may depend upon the number of bits assigned thereto. For
example, 32 bits or four octets may be assigned to the
identification value and thus the identification value would range
from 0 to 2.sup.31. With respect to the type key-value pair, the
type value is typically either request, end-request, response,
and/or any application-defined value. Any other suitable
application-defined key-value pairs may also be includes in the
service packet.
An exemplary service packet may be: <service type "request"
version "1.0" ID="1111" method="get" href
"http:/domain.com/whatever" acceptprotocol="http"/>
FIG. 5 is a flowchart illustrating a typical process 180 of a
peering service server in processing a request from a peering
client in the peer-to-peer network. At step 182, the service server
is listening on a designated port for a broadcast request message
from a peer client on the network. At step 184, the service server
receives a broadcast request message on the designated port such as
an "I need" packet from a peering client on the network. At step
186, the service server determines if it has a local aliased copy
of the requested item. In particular, the service server refers to
its list of local aliased copies to determine if the service server
has a local version of the requested resource or item, such as an
URL/URI.
If the service server determines that it does not have a local copy
of the requested resource, then the process 180 is complete.
Alternatively, if the service server determines that it has a local
copy, the service server preferably waits a randomly generated
delay response time period while listening for a broadcast "I
found" packet from the requesting client corresponding to the
received request packet at step 188. As discussed above, the
service server may generate a random number between 0 and 2000 as
the length of time in milliseconds it waits before responding. The
broadcast "I found" packet from the requesting client indicates
that the requesting client has found the requested resource.
It is noted that throughout the process 180, the service server is
preferably continuously listening for any additional broadcast
request messages and performs process 180 for each received
broadcast request message as they are received.
If the service server receives an "I found" packet from the
requesting client before expiry of the delay response time period,
the service server cancels the response to the request at step 190
and the process 180 is complete. Alternatively, if no "I found"
packet is received prior to the expiration of the delay response
time period, the service server transmits an "I have" packet
directly to the requesting peer client at step 192 and the server
process 180 is complete. The "I have" packet preferably contains a
local alias for the requested object on the service server.
FIG. 6 is a flowchart illustrating a typical process 200 of a
peering service client in requesting a resource over the
peer-to-peer network. At step 202, the service client generates and
broadcasts an "I need" packet over the peer-to-peer network on a
designated port. At step 204, the service awaits for a response
from any of the service servers on the network for a period equal
to a client response waiting time period, typically a maximum delay
response time period plus a transmission time period.
If an "I have" response is received from a service server during
the client response waiting time, then the service client generates
and broadcasts an "I found" message over the network at step 206.
The service-enabled application requesting the item then retrieves
the requested item from the responding service server at the
location within the network as specified in the received "I have"
packet at step 208. Once the service-enabled application
successfully retrieves the requested item, the service client
informs the service server running on the same machine that a local
copy of the resource now exists at step 210. The process 200 is
then complete.
Alternatively, if no response is received during the client
response waiting time, i.e., the service client times out, the
service client determines if the service-enabled application can or
will wait for completion of any in-progress download of the
requested item by another peer at step 214. If not, the service
client retrieves the requested item itself such as via the Internet
at step 216 and then proceeds to step 210 to complete the process
200.
If the service client determines that it can or will wait for the
completion of any in-progress download, the service client
determines whether there are any in-progress downloads at step 220.
If there are no such in-progress downloads of the requested item,
the service client then proceeds to step 210 to complete the
process 200.
If there is at least one in-progress download of the requested
item, then the service client generates and broadcasts an "I found"
message at step 222. The service client then awaits completion of
the in-progress download of the requested item at step 224. For
example, the service client may receive an "I have" or a "Download
complete" message from the downloading peer.
Upon completion of the in-progress download of the requested item,
the service client retrieves the requested item from the local
location within the network at step 226. After successful
completion of its request, the service client then proceeds to step
210 to complete the process 200. It is noted that steps 214 and 220
226 can be optional and preferably implemented for applications
that include downloading of files.
As is evident, in order for the service client to determine if
there are any in-progress downloads at step 220, a service client
that is downloading a file for a service-enabled application from
outside of the network, e.g., from the Internet, notifies its peers
on the network that a downloading process is in progress. For
example, FIG. 7 is a flowchart illustrating a preferred embodiment
of the retrieve step 216 in more detail.
As shown, the service client begins retrieving the requested item
at step 216A. At step 216B, the service client may broadcast a
"downloading" message and/or directly respond with a "I am
downloading" response message to any client server that transmitted
a broadcast "I need" request. In addition, the service client
preferably also periodically transmits progress packets at step
216C either by broadcast or by direct transmission to any node that
is waiting for the resource such that those nodes may interactively
display download progress information to end users at those nodes.
Alternatively, steps 216B and 216C may be combined into a single
periodic packet transmission in which each packet is a
"downloading" message that includes progress information.
Service-Enabled Product Updating Application
One exemplary implementation of the peering service described
herein is a product updating service implementation and a
service-enabled application having a shared agent. The agent is
shared by an anti-virus application and a firewall application. The
peering service is encapsulated in a single DLL that contains
components for performing an update service, namely, a peering
server having an HTTP server, a peering client, and a product
updating service.
The product updating service determines what updates, if any, to
request. If the product updating service determines that an update
is necessary, the service client broadcasts an "I need" packet to
request a specific URL for the necessary product updates. In other
words, the peering service provides a mechanism for keeping
service-enabled application, its engine, and its virus signature
files up-to-date.
In particular, when a first computer or node boots, its product
updater broadcasts an "I need" packet requesting for myupdate.cab
file at a specified URL. The myupdate.cab file, e.g., approximately
7 8k in size, contains a script with instructions on how to check
the current version of the product, engine, and virus signature
files against the latest available version so that the product
updater can determine if an update is necessary. This file may not
be cacheable, so the service servers may not be able to offer it
and can instead be obtained directly via the Internet.
If the product updating service determines, based on the
myupdate.cab file, that an update is necessary, the product
updating service, via the peering service, broadcasts an "I need"
packet over the network. An update may include engine, DAT, and/or
product updates. For any update files that are downloaded, whether
directly from the Internet and/or from one or more of the peers on
the network, the product updating service preferably checks to
ensure that the updates have been digitally signed. Once the
updates are authenticated, they are installed at the requesting
node.
The product update service checks for updates at any suitable
pre-defined intervals and/or upon occurrence of various events. For
example, the product update service may check for updates upon boot
or 5 minutes after boot, 6 hours after each unsuccessful check,
and/or upon a scheduled basis such as once a day, once every 12
hours after each successful check.
An update can include virus signature files (DATs), engine, and/or
product update. DATs are typically updated weekly, such on a
particular day of the week and are approximately 900 950k in size
on average. The engine is usually updated every 2 to 3 months and
is approximately 550 600k in size on average. The product is
updated as hotfixes become available, typically every 6 8 weeks, or
as new versions become available, typically every 4 6 months, and
is approximately 700 750k in size on average.
In the current example, a complete update, including engine, virus
signature files and product, comprises of six *.cab files, totaling
approximately 2.25M. The six *.cab files for an update and their
respective average sizes are listed below:
TABLE-US-00001 Myavdat.YYMMDDHHMM.cab average 910k
Myxtrdat.YYMMDDHHMM.cab average 16k Mycioagt.YYMMDDHHMM.cab average
370k Vsasap.YYMMDDHHMM.cab average 360k Vseng9x.YYMMDDHHMM.cab
average 240k Vsengine.YYMMDDHHMM.cab average 340k
As any number of these *.cab files may need to be updated, each
file is preferably requested via the peering service in a separate
transaction. Thus, some or all of the needed *.cab file may be
pulled from different nodes and/or the Internet.
Distributed Function Discovery Process Involving a Designated
Service Provider
In certain circumstances, an actual service provider in a
peer-to-peer network may not be easily locatable by other peers in
the network. In particular, because there is no single dedicated
server in a peer-to-peer network, a node attempting to locate a
service provider may not be able to locate the desired service
provider or may unnecessarily expend efforts and utilize resources
to locate the desired service provider. Thus, a distributed
function discovery process involving a designated service provider
is preferably implemented in order for peers to efficiently and
effectively discover the designated service providers in a
distributed manner within a peer-to-peer network environment.
A designated service provider may be elected, for example, as
described in co-pending U.S. patent application Ser. No.
09/921,941, entitled "System and Method for Automatic Selection of
Service Provider for Efficient use of Bandwidth and Resources In a
Peer-to-Peer Network Environment", filed on evendate herewith and
incorporated in its entirety by reference herein. Alternatively, a
designated service provider may simply be designated or selected in
any suitable manner.
In particular, under certain circumstances and/or for certain
applications or tasks, it may be desirable to have a designated
single point of responsibility such as by determining which
computer or node on the network is most suitable to perform the
desired service or task as a service provider. That computer thus
preferably becomes the designated service provider to which all
other peers on the peer-to-peer network will go to first for the
specified service.
Such a selection of a designated service provider may be useful,
for example, where some nodes are restricted from accessing the
Internet (as shown in FIG. 1). However, many applications and/or
their updates such as anti-virus and firewall applications are
delivered and administered via the Internet such as anti-virus
signature files. Thus, it may be preferably to designate a service
provider at a node with Internet access in a peer-to-peer network
to frequently check for and download files such as updates via the
Internet to ensure that all nodes can obtain the most up-to-date
versions of the distributable files or resources. Because peers on
the network share a LAN path, the non-connected nodes thus have
indirect access to the desired resources. In other words, an
Internet-connect machine may serve as the designated service
provider to act on behalf of nodes that are not connected to the
Internet. Preferably, each node implements a distributed function
discovery process involving a designated service provider in order
for peers to efficiently and effectively discover the designated
service provider in a distributed manner within a peer-to-peer
network environment.
FIG. 8 is a flowchart illustrating an exemplary process 400
implemented by a node for storing or caching designated service
provider information as part of a distributed function discovery
process in a peer-to-peer network. As shown, as step 402, the node
receives a packet containing information regarding the designated
service provider. At step 404, the node stores or caches the
designated service provider information to be served to other peers
on the network. At step 406, the node provides designated service
provider information as a server resource to other peers on the
network.
FIG. 9 is a flowchart illustrating an exemplary process 410
implemented by a node for serving stored or cached designated
service provider information as part of a distributed function
discovery process in a peer-to-peer network. It is noted that
process 410 is similar to process 180 of a peering service server,
as shown in and described above with reference to FIG. 5, in
processing a request for a resource such as a file from a peering
client in the peer-to-peer network.
As shown, at step 412, a peering server listens for a broadcast
request message from a peering client. At step 414, the peering
server receives a broadcast request message from the peering client
requesting a service function for which the peering server is not
the provider. At step 416, the peering server determines whether it
has stored or cached information regarding the designated service
provider corresponding to the requested service. The information
generally relates to the remote location at which the designated
service function provider is located.
If the peering server does not have such information stored or
cached, then the process 410 is complete. Alternatively, if the
peering server does have information stored or cached regarding the
designated service provider, then the peering server awaits a
randomly generated delay response period and listens for a
broadcast "I found" or similar packet from the requesting peering
client at step 418.
If the peering server receives such an "I found" or similar packet,
then the peering server cancels the request at step 420 and the
process 410 is complete. Alternatively, if the no such "I found" or
similar packet is received during the randomly generated delay
response period, then the peering server transmits an "I have" or
similar packet to the requesting peering client at step 422 and the
process 410 is complete. Preferably, the response at step 422 is
digitally signed, such as by a 1024-bit VeriSign digital
certificate. The "I have" packet would typically contain the remote
location information regarding the designated service provider such
that the requesting peering client is guided to the designated
service provider for the requested service function.
As is evident, a peer client may request a service function by
broadcasting a service function request packet over the network and
awaiting a response from a peer server. The response would contain
information regarding a designated provider for the requested
service function including a location of the designated provider
that is remote to the responding peer server. In addition, the peer
client would access the requested service function from the
designated service provider at the remote location specified in the
response of the responding peer server.
Thus, as is evident, by implementing the distributed function
discovery process at each node, those nodes can direct and guide
other service-requesting nodes to the desired designated service
provider.
As illustrated in the description above, the peering service
facilitates in reducing or minimizing the number of service clients
that have to obtain files or other resources such as product update
files via the Internet by using secure, peer-to-peer communication
to distribute the files among client machines on a network, such as
a LAN, via an intranet. The peering service enables secure,
automatic distribution of the update files between service clients,
independent of a network administrator or end-user, to keep the
anti-virus and firewall application/service up-to-date with minimal
impact to network bandwidth.
Often, many computers on a network do not have the most up-to-date
anti-virus and/or firewall files. Using the secure peering service
allows for automatic and secure updating of such files and also
reduces or eliminates the need for a network administrator to
script anti-virus file updates. Furthermore, by efficiently
spreading load and utilizing resources across a local network over
a high-speed LAN, a bandwidth crunch resulting from the computers
pulling update files from the Internet is largely reduced. Thus,
the peering service allows for ease of distribution of product
upgrades and updates with a minimal number of computers requiring
to connect to the Internet to obtain the necessary files resulting
in a reduced usage of Internet bandwidth.
The peering service also allows a given client to pull the data
files from any node on the network, rather than having to connect
to a centralized server that might require several additional
network hops, resulting in an optimal use of network bandwidth to
distribute updates.
FIGS. 10 and 11 illustrate a schematic and a block diagram,
respectively, of an example of a general purpose computer system
1000 suitable for executing software programs that implement the
methods and processes described herein. The architecture and
configuration of the computer system 1000 shown and described
herein are merely illustrative and other computer system
architectures and configurations may also be utilized.
The illustrative computer system 1000 includes a display 1003, a
screen 1005, a cabinet 1007, a keyboard 1009, and a mouse 1011. The
mouse 1011 can have one or more buttons for interacting with a GUI
(graphical user interface) that may be displayed on the screen
1005. The cabinet 1007 typically house one or more drives to read a
computer readable storage medium 1015, system memory 1053, and a
hard drive 1055, any combination of which can be utilized to store
and/or retrieve software programs incorporating computer codes that
implement the methods and processes described herein and/or data
for use with the software programs, for example. Examples of
computer or program code include machine code, as produced, for
example, by a compiler, or files containing higher level code that
may be executed using an interpreter.
Computer readable media may store program code for performing
various computer-implemented operations and may be encompassed as
computer storage products. Although a CD-ROM and a floppy disk 1015
are shown as exemplary computer readable storage media readable by
a corresponding CD-ROM or floppy disk drive 1013, any other
combination of computer readable storage media can be utilized.
Computer readable medium typically refers to any data storage
device that can store data readable by a computer system. Examples
of computer readable storage media include tape, flash memory,
system memory, and hard drive may alternatively or additionally be
utilized. Computer readable storage media may be categorized as
magnetic media such as hard disks, floppy disks, and magnetic tape;
optical media such as CD-ROM disks; magneto-optical media such as
floptical disks; and specially configured hardware devices such as
application-specific integrated circuits (ASICs), programmable
logic devices (PLDs), and ROM and RAM devices. Further, computer
readable storage medium may also encompass data signals embodied in
a carrier wave, such as the data signals embodied in a carrier wave
carried in a network. Such a network may be an intranet within a
corporate or other environment, the Internet, or any network of a
plurality of coupled computers such that the computer readable code
may be stored and executed in a distributed fashion.
Computer system 1000 comprises various subsystems. The subsystems
of the computer system 1000 may generally include a microprocessor
1051, system memory 1053, fixed storage 1055 (such as a hard
drive), removable storage 1057 (such as a CD-ROM drive), display
adapter 1059, sound card 1061, transducers 1063 (such as speakers
and microphones), network interface 1065, and/or scanner interface
1067.
The microprocessor subsystem 1051 is also referred to as a CPU
(central processing unit). The CPU 1051 can be implemented by a
single-chip processor or by multiple processors. The CPU 1051 is a
general purpose digital processor which controls the operation of
the computer system 1000. Using instructions retrieved from memory,
the CPU 1051 controls the reception and manipulation of input data
as well as the output and display of data on output devices.
The network interface 1065 allows CPU 1051 to be coupled to another
computer, computer network, or telecommunications network using a
network connection. The CPU 1051 may receive and/or send
information via the network interface 1065. Such information may
include data objects, program instructions, output information
destined to another network. An interface card or similar device
and appropriate software implemented by CPU 1051 can be used to
connect the computer system 1000 to an external network and
transfer data according to standard protocols. In other words,
methods and processes described herein may be executed solely upon
CPU 1051 and/or may be performed across a network such as the
Internet, intranet networks, or LANs (local area networks), in
conjunction with a remote CPU that shares a portion of the
processing. Additional mass storage devices (not shown) may also be
connected to CPU 1051 via the network interface 1065.
The subsystems described herein are merely illustrative of the
subsystems of a typical computer system and any other suitable
combination of subsystems may be implemented and utilized. For
example, another computer system may also include a cache memory
and/or additional processors 1051, such as in a multi-processor
computer system.
The computer system 1000 also includes a system bus 1069. However,
the specific buses shown are merely illustrative of any
interconnection scheme serving to link the various subsystems. For
example, a local bus can be utilized to connect the central
processor to the system memory and display adapter.
While the preferred embodiments of the present invention are
described and illustrated herein, it will be appreciated that they
are merely illustrative and that modifications can be made to these
embodiments without departing from the spirit and scope of the
invention. Thus, the invention is intended to be defined only in
terms of the following claims.
* * * * *
References